The consumption of enormous amounts of food characteristic of Prader- Willi syndrome (PWS) not only leads to obesity, but also to severe behavior problems and other forms of psychosocial dysfunction. Considerable evidence suggests serotonin or endogenous opioids or both may be involved in the excessive eating seen in PWS. The present project will investigate effects of fluoxetine ( a serotonin reuptake blocker) and naltrexone (an opiate antogonist) on food preference and food motivation. It is possible these drugs may affect desirability of foods as a function of their macroconstituents, rather than affecting hunger, as such. A progressively demanding exercise procedure rewarded by small amounts of food prepared with specific macroconsituent ratios will be used to assess the motivational properties of various food commodities under placebo, fluoxetine and naltrexone conditions. Each unit of exercise will require a progressively greater expenditure of energy. Food motivation will be measured by the total units of exercise completed each session. The point at which the subject stops in a session is called the 'break point,' and reflects the subject's motivation to eat in relation to the type of food available during that session. The progressive ratio procedure is a safer method of assessing food motivation than uncontrolled access to food, which can lead to dangerous levels of weight gain in a relatively short period. The results may suggest underlying neurochemical abnormalities which are regulated by specific defects on chromosome 15. An effort will be made to relate the results obtained in this project to those of metabolic and nutritional assessment, the Psychobehavioral Core, and the Genetic and Clinical Core. Of particular interest are relations between genetic subtypes of PWS, food motivation, and metabolic, neurochemical and behavioral patterns.